Theoretical Underpinnings

1 https://thehill.com/homenews/administration/376408-trump-if-you-dont-have-steel-you-dont-have-a-country/ 2 https://www.wsj.com/politics/national-security/scale-of-chinese-spying-overwhelms-western-governments-6ae644d2 The Roots of Economic Discontent Lie in the Dollar

The Triffin World

The deep unhappiness with the prevailing economic order is rooted in persistent overvaluation of the dollar and asymmetric trade conditions. Such overvaluation makes U.S. exports less competitive, U.S. imports cheaper, and handicaps American manufacturing. Manufacturing employment declines as factories close. Those local economies subside, many working families are unable to support themselves and become addicted to government handouts or opioids or move to more prosperous locations. Infrastructure declines as governments no longer service it, and housing and factories lay abandoned. Communities are “blighted.”

According to Autor, Dorn and Hanson (2016), between 600,000 and one million U.S. manufacturing jobs disappeared between 2000 and 2011 due to the “China shock” of increased trade with China. Including broader categories, the jobs displaced by trade during that decade were closer to 2 million. Even 2 million job losses over a decade represents only 200,000 per year, a fraction of the churn of jobs that occurs every year because of technology, rising and falling firms and sectors, and the economic cycle.

But that logic was flawed in two ways: first, the estimates of job losses due to trade increased over time as new research emerged, for instance Autor, Dorn and Hanson (2021); the “China shock” was much larger than initially estimated. Indeed, plenty of nonmanufacturing jobs which depended on local manufacturing economies were lost as well. Second, many job losses were concentrated in states and specific towns where alternative employment was not easily available. For these communities, the losses were severe.

The problem is compounded by the reversal of “the end of history” and the return of national security threats. With no major geopolitical rivals, U.S. leaders believed they could minimize the significance of declining industrial plant. But with China and Russia as not only trade but security threats, having a robust and well diversified manufacturing sector is of renewed necessity. If you have no supply chains with which to produce weapons and defense systems, you have no national security.

As President Trump argued, “if you don’t have steel, you don’t have a country.”1 While many economists fail to include such externalities in their analysis and are therefore happy to rely on trade partners and allies for such supply chains, the Trump camp does not share that trust. Many of America’s allies and partners have significantly larger trade and investment flows with China than they do with America; are we so sure we can trust them, if worse comes to worst?

Such problems are compounded by aggressive Chinese espionage. According to reports in the Wall Street Journal,2 in September alone, “the Federal Bureau of Investigation said a Chinese state-linked firm hacked 260,000 internet-connected devices, including cameras and routers, in the U.S., Britain, France, Romania and elsewhere [and] a Congressional probe said Chinese cargo cranes used at U.S. seaports had embedded technology that could allow Beijing to secretly control them.” The security, espionage and sabotage vulnerability of sensitive imports from China continues to grow.

In this telling, persistent overvaluation of the dollar is the key mechanism for trade imbalances, keeping imports from abroad stubbornly cheap despite widening trade deficits. So how is it possible that currency markets, which are the largest markets in the world in terms of sheer trading volume, don’t equilibrate?

The answer lies in the fact that there are (at least) two concepts of equilibrium for currencies. One is rooted in models of international trade. In trade models, currencies adjust over the long term to balance international trade. If a country

Share of total employment Manufacturing employees MA N U FACTU R I N G E M PLOYM E NT 5,000 0 10,000 15,000 20,000 1940 1950 1960 1970 1980 1990 2000 2010 2020 10% 0% 20% 30% 40%

Figure 1 Manufacturing employment in the United States. Source: Bureau of Labor Statistics runs a trade surplus for a sustained period, it receives foreign currency for its goods, which it then sells for its domestic currency, pushing its domestic currency higher. That process occurs until its currency is strong enough that its exports decline and imports increase, balancing trade.

The other equilibrium concept is financial and comes from savers selecting investment alternatives among different nations. In this equilibrium concept, currencies adjust to make investors indifferent between holding assets denominated in different currencies, on an ex ante risk-adjusted basis.

However, the latter class of models becomes more complicated when a nation’s currency is a reserve asset, as is America’s. Because America provides reserve assets to the world, there is demand for U.S. dollars (USD) and U.S. Treasury securities (USTs) that is not rooted in balancing trade or in optimizing risk-adjusted returns.

These reserve functions serve to facilitate international trade and provide a vehicle for large pools of savings, often held for policy reasons (e.g. reserve or currency management or sovereign wealth funds) rather than return maximization.

Much (but not all) of the reserve demand for USDs and USTs is inelastic with respect to economic or investment fundamentals. Treasurys bought to collateralize trade between Micronesia and Polynesia are bought irrespective of the U.S. trade balance with either, the latest jobs report, or the relative return of Treasurys vs. German Bunds. Such phenomena reflect what can be described as a “Triffin world,” after Belgian economist Robert Triffin.

In Triffin world, reserve assets are a form of global money supply, and demand for them is a function of global trade and savings, not the domestic trade balance or return characteristics of the reserve nation.

When the reserve country is large relative to the rest of the world, there are no significant externalities imposed on the reserve country from its reserve status. The distance from the Triffin equilibrium to the trade equilibrium is small. However, when the reserve country is smaller relative to the rest of the world—say, because global growth exceeds the reserve country’s growth for a long period of time—tensions build and the distance between the Triffin equilibrium and the trade equilibrium can be quite large. Demand for reserve assets leads to significant currency overvaluation with real economic consequences.

In Triffin world, the reserve asset producer must run persistent current account deficits as the flip side of exporting reserve assets. USTs become exported products which fuel the global trade system. In exporting USTs, America receives foreign currency, which is then spent, usually on imported goods. America runs large current account deficits not because it imports too much, but it imports too much because it must export USTs to provide reserve assets and facilitate global growth. This view has been discussed by prominent policymakers from both the United States (e.g. Feldstein and Volcker, 2013) as well as China (e.g. Zhou, 2009).3

As the United States shrinks relative to global GDP, the current account or fiscal deficit it must run to fund global trade and savings pools grows larger as a share of the domestic economy. Therefore, as the rest of the world grows, the consequences for our own export sectors—an overvalued dollar incentivizing imports—become more difficult to bear, and the pain inflicted on that portion of the economy increases.

Eventually (in theory), a Triffin “tipping point” is reached at which such deficits grow large enough to induce credit risk in the reserve asset. The reserve country may lose reserve status, ushering in a wave of global instability, and this is referred to as the Triffin “dilemma.” Indeed, the paradox of being a reserve currency is that it leads to permanent twin deficits, which in turn lead over time to an unsustainable accumulation of public and foreign debt that eventually undermines the safety and reserve currency status of such large debtor economy.

While the United States share of global GDP halved from 40 percent of global GDP in the 1960s to 21 percent in 2012, and has recovered slightly to its present level of 26%, it is still far from such a tipping point, in part because there are no meaningful alternatives to the dollar or the UST. A reserve currency must be convertible into other currencies, and a reserve asset must be a stable store of value governed by reliable rule of law. While other nations like China aspire to reserve status, they satisfy neither of these criteria. And while Europe may, its bond markets are fragmented relative to the UST market, and its share of global GDP has shrunk even more than America’s. It is worth observing that the U.S. share of global GDP troughed around the GFC and has stabilized or improved since then, coincident with the pattern in manufacturing employment.

In this telling, our share of global GDP drives the size of the Triffin distortion in trade equilibrium, which in turn drives the state of the tradeable sector. The backdrop for these currency developments has been a system of tariff rates defining the international trading system that are, broadly speaking, locked into a configuration designed for a different economic age. According to the World Trade Organization, the United States effective tariff on imports is the lowest any nation in the world imposes at about 3%, while the European Union imposes about 5% and China 10%.4 These numbers are averages across all imports and not reflective of bilateral tariff rates; bilateral discrepancies can be much larger, for instance the U.S. imposes only 2.5% tariffs on auto imports from the E.U., while Europe imposes a 10% duty on American auto imports.5 Many developing nations apply much higher rates, and Bangladesh has the world’s highest effective rate at 155%. These tariffs are, in large part, legacies of an era in which the United States wanted to generously open its markets to the rest of the world at advantageous terms to assist with rebuilding after World War II, or in creating alliances during the Cold War. Moreover, tariffs in 3 A critical analysis of the “current account” and “fiscal” versions of the Triffin dilemma is available in Bordo and McCauley (2017).

Their dismissal of these theories focuses more on the lack of dilemma/crisis in the near term and the inability to identify any such crossing point, rather than on an ability to discount the basic mechanism. 4 https://www.wto.org/english/res_e/booksp_e/world_tariff_profiles24_e.pdf 5 https://www.politico.com/news/2023/01/19/joe-manchin-davos-inflation-reduction-act-europe-00078510

some cases enormously understate the unevenness of the playing field, as some nations employ material nontariff barriers, steal intellectual property, and more. In theory, prior tariff rates may not affect trade if floating currencies adjust to offset them, but they have very significant consequences for revenue and burden sharing (see the discussion below). Figure 2 U.S. share of global GDP. Source: World Bank Economic Consequences While we are likely far from the economic crises that comprise the tipping point of the Triffin dilemma, we must nonetheless reckon with the consequences of the Triffin world. Reserve nation status comes with three major consequences: somewhat cheaper borrowing, more expensive currency, and the ability pursue security goals via the financial system.

1. Cheaper Borrowing Because there is a persistent reserve-driven demand for UST securities, the United States may be able to borrow at lower yields than would otherwise be the case. Because economists have little variation to study (we’ve been the sole reserve currency for many decades), it’s impossible to have confidence about how big this benefit is. Some estimates, fictional as they are, have it as large as 50-60 basis points of borrowing yield (McKinsey, 2009). In any case, there are numerous countries that borrow significantly more cheaply than the United States. At the time of writing, all G7 members borrow more cheaply than the United States except for the UK, which borrows a tenth of a percent more expensive. Other peers like Switzerland and Sweden borrow more cheaply too, Switzerland by almost 4 percentage points. Meanwhile, an erstwhile troubled debtor like Greece can borrow over a point more cheaply. A User’s Guide to Restructuring the Global Trading System 9 More precisely, one can create a synthetic dollar borrowing rate with currency risk hedged out, i.e. examine deviations from covered interest parity, as in Du, Im and Schreger (2018). Such deviations are currently (and usually) close to zero6 for the United States relative to other G10 borrowers; in other words, there’s little special borrowing rate conferred on the U.S. relative to other developed countries. G10 vs. EM, however, still contain substantial residuals, suggesting that emerging markets pay a borrowing premium relative to developed markets to borrow. The inference I draw from this is that while, all else equal, being a reserve currency may reduce borrowing costs, whatever benefit is gained is likely to be dwarfed by things like central bank policy outlooks, growth and inflation forecasts, and equity market performance. However, the borrowing advantage may be framed differently: rather than reducing the cost of borrowing, it may reduce the price sensitivity of borrowing. In other words, we don’t necessarily borrow substantially cheaper, but we can borrow more without pushing yields higher. This is a consequence of the price inelasticity of demand for reserve assets, and the flip side that we run large external deficits to finance that reserve provision. 6 https://sites.google.com/view/jschreger/CIP?authuser=0

2. Richer Currency

The more significant macroeconomic consequence of serving as the world’s reserve producer is that reserve demand for American assets pushes up the dollar, leading it to levels far in excess of what would balance international trade over the long run. According to the IMF, there are about $12 trillion of global foreign exchange reserves in official hands, of which roughly 60% are allocated in dollars—in reality, reserve holdings of the dollar are much higher, as quasi- and non-official entities hold dollar assets for reserve purposes too. Clearly, $7 trillion of demand is enough to move the needle in any market, even currency markets. For reference, $7 trillion is roughly a third of U.S. M2 money supply; flows creating or unwinding these holdings will obviously have significant market consequences. If trillions of dollars of securities bought for the Fed’s policy and not investment purposes on the Fed’s balance sheet have had any effect on financial markets, then trillions of dollars bought for foreign central banks’ policies and not investing purposes should also have some effect if they are on other nations’ balance sheets instead of the Fed’s. Because nations accumulate reserves in part to stem appreciation pressures in their own currencies, there is a contemporaneous negative correlation between the exchange value of the dollar and the level of global reserves. Reserves tend to go up when the dollar is going down, as accumulators buy dollars to suppress their currencies, and vice versa when the dollar is going up. Nevertheless, other than two quarters in 1991, the United States has run a current account deficit since 1982. That the current account cannot balance beyond a negligible period over half a century tells us the dollar is not playing its role of equilibrating international trade and income flows. The interplay between reserve status and the loss of manufacturing jobs is sharpest during economic downturns. Because the reserve asset is “safe,” the dollar appreciates during recessions. By contrast, other nations’ currencies tend to depreciate when they go through an economic downturn. That means that when aggregate demand suffers a decline, pain in export sectors get compounded by a sharp erosion of competitiveness. Thus employment in manufacturing declines steeply during a recession in the United States, and then fails to recover materially afterward.

Canada -1.05 Japan -3.38 UK 0.12 France -1.19 Germany -1.94 Italy -0.66 Greece -1.03 Switzerland -3.93 Sweden -2.20

Table 1: 10-year borrowing spreads to Treasury notes. Negative numbers mean the other nation borrows more cheaply than the United States. Source: Bloomberg, HBC calculations A User’s Guide to Restructuring the Global Trading System 10

CU RRE NT ACC OU NT 1961 1950 1960 1970 1980 1990 2000 2010 2020 2024 -1200 -1050 -900 -750 -600 -450 -300 -150 150 300 0 -8% -7% -6% -5% -4% -3% -2% -1% 1% 2% 0% Current account (billions of $) Current account (% G DP) Figure 3 U.S. current account. Source: Bureau of Economic Analysis, HBC calculations It may seem odd to suppose that reserve demand for Treasury securities plays only a small role in delivering favorable borrowing terms, yet a large role in creating currency overvaluation. However, that is the explanation most consistent with outcomes in both interest rate markets and the balance of payments. Indeed, it is also consistent with the idea that liquidity injections ultimately raise interest rates because they stimulate stronger nominal growth. Other theoretical explanations for this combination of outcomes could make interesting research.

3. Financial Extraterritoriality

Finally, if the reserve asset is the lifeblood of the global trade and financial systems, it means that whoever controls the reserve asset and currency can exert some level of control trade and financial transactions. That allows America to exert its will in foreign and security policy using financial force instead of kinetic force. America can, and does, sanction people all over the globe in a variety of ways. From freezing assets to cutting nations off from SWIFT and restricting access to the U.S. banking and financial system that is critical for any foreign bank doing global business, the U.S. exerts its financial might to achieve foreign policy ends of weakening enemies without having to mobilize a single soldier. Economists cannot evaluate whether America’s national security goals are worthy, only note that it can achieve them far more cheaply because of America’s control of the international trade and financial systems by virtue of our reserve currency status.

A comprehensive review of how the United States mobilizes the global financial architecture for national security purposes is given in Cipriani, Goldberg and La Spada (2023), and a history of many of the key players in Mohsin (2024). In a broader sense, sanctions can also be perceived as a modern-day form of a blockade. Many previous reserve providing nations possessed significant sea power by virtue of their trading empires, which allowed them A User’s Guide to Restructuring the Global Trading System 11 to blockade rival nations and hamper their economic production; sanctions achieve related outcomes, without the need for physical action.

The Core Tradeoff

Synthesizing these properties of reserve assets, if there is persistent, price-inelastic demand for reserve assets but only modestly cheaper borrowing, then America’s status as reserve currency confers the burden of an overvalued currency eroding the competitiveness of our export sector, balanced against the geopolitical advantages of achieving core national security aims at minimal cost via financial extraterritoriality. The tradeoff is thus between export competitiveness and financial power projection. Because power projection is inextricable from the global security order America underwrites, we need to understand the question of reserve status as intertwined with national security. America provides a global defense shield to liberal democracies, and in exchange, America receives the benefits of reserve status—and, as we are grappling with today, the burdens. This connection helps explain why President Trump views other nations as taking advantage of America in both defense and trade simultaneously: the defense umbrella and our trade deficits are linked, through the currency. In a Triffin world, this arrangement becomes more challenging as the United States shrinks as a share of global GDP and military might. As the economic burdens on America grow with global GDP outpacing American GDP, America finds it more difficult to underwrite global security, because the current account deficit grows and our ability to produce equipment becomes hollowed out. The growing international deficit is a problem because of the increased strain it places on the American export sector and the socioeconomic problems that follow therefrom. That the bargain becomes less appealing in this context brings us to present, whereby there is growing consensus in America to change the relationship.

Reshaping the Global System

If America is unwilling to bear the status quo, then it will take steps to change it. There are, broadly speaking, unilateral and multilateral approaches, and approaches focused on tariffs or currencies. Unilateral solutions are more likely to have undesired side effects, like market volatility. Multilateral solutions may have less volatility, but entail the difficulty of getting trading partners onboard, which curtails the size of the potential gains from reshaping the system. Unilateral policies provide greater flexibility to rapidly shift policy; multilateral policies are more difficult (maybe impossible?) to implement, but allow you to recruit foreign policymakers to help reduce volatility.

The U.S. dollar is the reserve asset in large part because America provides stability, liquidity, market depth and the rule of law. Those are related to the characteristics that make America powerful enough to project physical force worldwide and allow it to shape and defend the global international order. The history of intertwinement between reserve currency status and national security is long. In any possible reshaping of the global trading system, these linkages will become ever more explicit.

Both tariffs and currency policy are aimed at improving the competitiveness of American manufacturing, and thus increasing our industrial plant and allocating aggregate demand and jobs from the rest of the world stateside. These policies are unlikely to result in significant reshoring of low-value-added industries like textiles, for which other countries—like Bangladesh—will retain comparative advantage despite significant swings in currency or tariff rates. However, these policies can help preserve the American edge in high-value-added manufacturing, slow down and prevent further offshoring, and potentially increase negotiating leverage with which to procure agreements from other countries to open their markets to American exports or protect American intellectual property rights. The Phase 1 trade deal with China in 2019 made advances in these domains, before China abdicated its commitments under that agreement.

Moreover, because many in the Trump camp see trade policy and national security as inextricably intertwined, many interventions will be targeted at industrial plant critical to security, to the extent they can. National security will likely become ever more broadly conceived, for instance to include products like semiconductors and pharmaceuticals.

Despite the dollar’s role in weighing heavily on the U.S. manufacturing sector, President Trump has emphasized the value he places on its status as the global reserve currency, and threatened to punish countries that move away from the dollar. I expect this tension to be resolved by policies that aim to preserve the status of the dollar, but improve burden sharing with our trading partners. International trade policy will attempt to recapture some of the benefit our reserve provision conveys to trading partners and connect this economic burden sharing with defense burden sharing. Although the Triffin effects have weighed on the manufacturing sector, there will be attempts to improve America’s position within the system without destroying the system. No matter what policy is adopted, there is the risk of material adverse consequences for financial markets and the economy. However, there are steps the Administration can take to try to mitigate these consequences and make the policy changes as successful as possible.